Examples with CovarianceMatrix edu.cmu.tetrad.data.CovarianceMatrix used on opensource projects

Example 6 with CovarianceMatrix

use of edu.cmu.tetrad.data.CovarianceMatrix in project tetrad by cmu-phil.

the class Mimbuild2 method search.

// =================================== PUBLIC METHODS =========================================//
public Graph search(List<List<Node>> clustering, List<String> latentNames, ICovarianceMatrix measuresCov) {
    List<String> _latentNames = new ArrayList<>(latentNames);
    List<String> allVarNames = new ArrayList<>();
    for (List<Node> cluster : clustering) {
        for (Node node : cluster) allVarNames.add(node.getName());
    }
    measuresCov = measuresCov.getSubmatrix(allVarNames);
    List<List<Node>> _clustering = new ArrayList<>();
    for (List<Node> cluster : clustering) {
        List<Node> _cluster = new ArrayList<>();
        for (Node node : cluster) {
            _cluster.add(measuresCov.getVariable(node.getName()));
        }
        _clustering.add(_cluster);
    }
    List<Node> latents = defineLatents(_latentNames);
    this.latents = latents;
    // This removes the small clusters and their names.
    removeSmallClusters(latents, _clustering, getMinClusterSize());
    this.clustering = _clustering;
    Node[][] indicators = new Node[latents.size()][];
    for (int i = 0; i < latents.size(); i++) {
        indicators[i] = new Node[_clustering.get(i).size()];
        for (int j = 0; j < _clustering.get(i).size(); j++) {
            indicators[i][j] = _clustering.get(i).get(j);
        }
    }
    TetradMatrix cov = getCov(measuresCov, latents, indicators);
    CovarianceMatrix latentscov = new CovarianceMatrix(latents, cov, measuresCov.getSampleSize());
    this.latentsCov = latentscov;
    Graph graph;
    Cpc search = new Cpc(new IndTestFisherZ(latentscov, alpha));
    search.setKnowledge(knowledge);
    graph = search.search();
    // try {
    // Ges search = new Ges(latentscov);
    // search.setDepErrorsAlpha(penaltyDiscount);
    // search.setKnowledge(knowledge);
    // graph = search.search();
    // } catch (Exception e) {
    // //            e.printStackTrace();
    // CPC search = new CPC(new IndTestFisherZ(latentscov, alpha));
    // search.setKnowledge(knowledge);
    // graph = search.search();
    // }
    this.structureGraph = new EdgeListGraph(graph);
    GraphUtils.fruchtermanReingoldLayout(this.structureGraph);
    return this.structureGraph;
}

Example 7 with CovarianceMatrix

use of edu.cmu.tetrad.data.CovarianceMatrix in project tetrad by cmu-phil.

the class MimbuildTrek method search.

// =================================== PUBLIC METHODS =========================================//
public Graph search(List<List<Node>> clustering, List<String> latentNames, ICovarianceMatrix measuresCov) {
    List<String> _latentNames = new ArrayList<>(latentNames);
    List<String> allVarNames = new ArrayList<>();
    for (List<Node> cluster : clustering) {
        for (Node node : cluster) allVarNames.add(node.getName());
    }
    measuresCov = measuresCov.getSubmatrix(allVarNames);
    List<List<Node>> _clustering = new ArrayList<>();
    for (List<Node> cluster : clustering) {
        List<Node> _cluster = new ArrayList<>();
        for (Node node : cluster) {
            _cluster.add(measuresCov.getVariable(node.getName()));
        }
        _clustering.add(_cluster);
    }
    List<Node> latents = defineLatents(_latentNames);
    this.latents = latents;
    // This removes the small clusters and their names.
    removeSmallClusters(latents, _clustering, getMinClusterSize());
    this.clustering = _clustering;
    Node[][] indicators = new Node[latents.size()][];
    for (int i = 0; i < latents.size(); i++) {
        indicators[i] = new Node[_clustering.get(i).size()];
        for (int j = 0; j < _clustering.get(i).size(); j++) {
            indicators[i][j] = _clustering.get(i).get(j);
        }
    }
    TetradMatrix cov = getCov(measuresCov, latents, indicators);
    CovarianceMatrix latentscov = new CovarianceMatrix(latents, cov, measuresCov.getSampleSize());
    this.latentsCov = latentscov;
    Graph graph;
    Cpc search = new Cpc(new IndTestTrekSep(measuresCov, alpha, clustering, latents));
    search.setKnowledge(knowledge);
    graph = search.search();
    // try {
    // Ges search = new Ges(latentscov);
    // search.setCorrErrorsAlpha(penaltyDiscount);
    // search.setKnowledge(knowledge);
    // graph = search.search();
    // } catch (Exception e) {
    // //            e.printStackTrace();
    // CPC search = new CPC(new IndTestFisherZ(latentscov, alpha));
    // search.setKnowledge(knowledge);
    // graph = search.search();
    // }
    this.structureGraph = new EdgeListGraph(graph);
    GraphUtils.fruchtermanReingoldLayout(this.structureGraph);
    return this.structureGraph;
}

Example 8 with CovarianceMatrix

use of edu.cmu.tetrad.data.CovarianceMatrix in project tetrad by cmu-phil.

the class PerformanceTestsDan method testIdaOutputForDan.

private void testIdaOutputForDan() {
    int numRuns = 100;
    for (int run = 0; run < numRuns; run++) {
        double alphaGFci = 0.01;
        double alphaPc = 0.01;
        int penaltyDiscount = 1;
        int depth = 3;
        int maxPathLength = 3;
        final int numVars = 15;
        final double edgesPerNode = 1.0;
        final int numCases = 1000;
        // final int numLatents = RandomUtil.getInstance().nextInt(3) + 1;
        final int numLatents = 6;
        // writeToFile = false;
        PrintStream out1;
        PrintStream out2;
        PrintStream out3;
        PrintStream out4;
        PrintStream out5;
        PrintStream out6;
        PrintStream out7;
        PrintStream out8;
        PrintStream out9;
        PrintStream out10;
        PrintStream out11;
        PrintStream out12;
        File dir0 = new File("gfci.output");
        dir0.mkdirs();
        File dir = new File(dir0, "" + (run + 1));
        dir.mkdir();
        try {
            out1 = new PrintStream(new File(dir, "hyperparameters.txt"));
            out2 = new PrintStream(new File(dir, "variables.txt"));
            out3 = new PrintStream(new File(dir, "dag.long.txt"));
            out4 = new PrintStream(new File(dir, "dag.matrix.txt"));
            out5 = new PrintStream(new File(dir, "coef.matrix.txt"));
            out6 = new PrintStream(new File(dir, "pag.long.txt"));
            out7 = new PrintStream(new File(dir, "pag.matrix.txt"));
            out8 = new PrintStream(new File(dir, "pattern.long.txt"));
            out9 = new PrintStream(new File(dir, "pattern.matrix.txt"));
            out10 = new PrintStream(new File(dir, "data.txt"));
            out11 = new PrintStream(new File(dir, "true.pag.long.txt"));
            out12 = new PrintStream(new File(dir, "true.pag.matrix.txt"));
        } catch (FileNotFoundException e) {
            e.printStackTrace();
            throw new RuntimeException(e);
        }
        out1.println("Num _vars = " + numVars);
        out1.println("Num edges = " + (int) (numVars * edgesPerNode));
        out1.println("Num cases = " + numCases);
        out1.println("Alpha for PC = " + alphaPc);
        out1.println("Alpha for FFCI = " + alphaGFci);
        out1.println("Penalty discount = " + penaltyDiscount);
        out1.println("Depth = " + depth);
        out1.println("Maximum reachable path length for dsep search and discriminating undirectedPaths = " + maxPathLength);
        List<Node> vars = new ArrayList<>();
        for (int i = 0; i < numVars; i++) vars.add(new GraphNode("X" + (i + 1)));
        // Graph dag = DataGraphUtils.randomDagQuick2(varsWithLatents, 0, (int) (varsWithLatents.size() * edgesPerNode));
        Graph dag = GraphUtils.randomGraph(vars, 0, (int) (vars.size() * edgesPerNode), 5, 5, 5, false);
        GraphUtils.fixLatents1(numLatents, dag);
        // List<Node> varsWithLatents = new ArrayList<Node>();
        // 
        // Graph dag = getLatentGraph(_vars, varsWithLatents, edgesPerNode, numLatents);
        out3.println(dag);
        printDanMatrix(vars, dag, out4);
        SemPm pm = new SemPm(dag);
        SemIm im = new SemIm(pm);
        NumberFormat nf = new DecimalFormat("0.0000");
        for (int i = 0; i < vars.size(); i++) {
            for (Node var : vars) {
                if (im.existsEdgeCoef(var, vars.get(i))) {
                    double coef = im.getEdgeCoef(var, vars.get(i));
                    out5.print(nf.format(coef) + "\t");
                } else {
                    out5.print(nf.format(0) + "\t");
                }
            }
            out5.println();
        }
        out5.println();
        String vars_temp = vars.toString();
        vars_temp = vars_temp.replace("[", "");
        vars_temp = vars_temp.replace("]", "");
        vars_temp = vars_temp.replace("X", "");
        out2.println(vars_temp);
        List<Node> _vars = new ArrayList<>();
        for (Node node : vars) {
            if (node.getNodeType() == NodeType.MEASURED) {
                _vars.add(node);
            }
        }
        String _vars_temp = _vars.toString();
        _vars_temp = _vars_temp.replace("[", "");
        _vars_temp = _vars_temp.replace("]", "");
        _vars_temp = _vars_temp.replace("X", "");
        out2.println(_vars_temp);
        DataSet fullData = im.simulateData(numCases, false);
        DataSet data = DataUtils.restrictToMeasured(fullData);
        ICovarianceMatrix cov = new CovarianceMatrix(data);
        final IndTestFisherZ independenceTestGFci = new IndTestFisherZ(cov, alphaGFci);
        final edu.cmu.tetrad.search.SemBicScore scoreGfci = new edu.cmu.tetrad.search.SemBicScore(cov);
        out6.println("GFCI.PAG_of_the_true_DAG");
        GFci gFci = new GFci(independenceTestGFci, scoreGfci);
        gFci.setVerbose(false);
        gFci.setMaxDegree(depth);
        gFci.setMaxPathLength(maxPathLength);
        // gFci.setPossibleDsepSearchDone(true);
        gFci.setCompleteRuleSetUsed(true);
        Graph pag = gFci.search();
        out6.println(pag);
        printDanMatrix(_vars, pag, out7);
        out8.println("Pattern_of_the_true_DAG OVER MEASURED VARIABLES");
        final IndTestFisherZ independencePc = new IndTestFisherZ(cov, alphaPc);
        Pc pc = new Pc(independencePc);
        pc.setVerbose(false);
        pc.setDepth(depth);
        Graph pattern = pc.search();
        out8.println(pattern);
        printDanMatrix(_vars, pattern, out9);
        out10.println(data);
        out11.println("True PAG_of_the_true_DAG");
        final Graph truePag = new DagToPag(dag).convert();
        out11.println(truePag);
        printDanMatrix(_vars, truePag, out12);
        out1.close();
        out2.close();
        out3.close();
        out4.close();
        out5.close();
        out6.close();
        out7.close();
        out8.close();
        out9.close();
        out10.close();
        out11.close();
        out12.close();
    }
}

Example 9 with CovarianceMatrix

use of edu.cmu.tetrad.data.CovarianceMatrix in project tetrad by cmu-phil.

the class FactorAnalysisAction method main.

public static void main(String[] args) {
    java.util.List<Node> nodes = new ArrayList<>();
    for (int i = 0; i < 9; i++) {
        nodes.add(new ContinuousVariable("X" + (i + 1)));
    }
    Graph graph = new Dag(GraphUtils.randomGraph(nodes, 0, 9, 30, 15, 15, false));
    SemPm pm = new SemPm(graph);
    SemIm im = new SemIm(pm);
    DataSet data = im.simulateData(500, false);
    ICovarianceMatrix cov = new CovarianceMatrix(data);
    FactorAnalysis factorAnalysis = new FactorAnalysis(cov);
    // factorAnalysis.centroidUnity();
    factorAnalysis.successiveResidual();
}

Example 10 with CovarianceMatrix

use of edu.cmu.tetrad.data.CovarianceMatrix in project tetrad by cmu-phil.

the class SemEstimatorWrapper method setParams.

private boolean setParams(SemPmWrapper semPmWrapper, DataModelList dataModel) {
    for (DataModel model : dataModel) {
        if (model instanceof DataSet) {
            DataSet dataSet = (DataSet) model;
            SemPm semPm = semPmWrapper.getSemPm();
            this.semPm = semPm;
            SemEstimator estimator = new SemEstimator(dataSet, semPm, getOptimizer());
            estimator.setNumRestarts(getParams().getInt("numRestarts", 1));
            estimator.setScoreType((ScoreType) getParams().get("scoreType", ScoreType.Fgls));
            if (!degreesOfFreedomCheck(semPm)) {
                return true;
            }
            estimator.estimate();
            getMultipleResultList().add(estimator);
        } else if (model instanceof ICovarianceMatrix) {
            ICovarianceMatrix covMatrix = new CovarianceMatrix((ICovarianceMatrix) model);
            SemPm semPm = semPmWrapper.getSemPm();
            this.semPm = semPm;
            SemEstimator estimator = new SemEstimator(covMatrix, semPm, getOptimizer());
            estimator.setNumRestarts(getParams().getInt("numRestarts", 1));
            estimator.setScoreType((ScoreType) getParams().get("scoreType", ScoreType.Fgls));
            if (!degreesOfFreedomCheck(semPm)) {
                return true;
            }
            estimator.estimate();
            getMultipleResultList().add(estimator);
        } else {
            throw new IllegalArgumentException("Data must consist of continuous data sets or covariance matrices.");
        }
    }
    return false;
}